Census (Hansard, 23 January 2003)

§ Ruth Kelly

The information requested falls within the responsibility of the National Statistician. I have asked him to reply.

Letter from L. Cook to Mr. E. Pickles, dated 23 January 2003: As National Statistician and Registrar General for England and Wales I have been asked to reply to your recent question to the Chancellor of the Exchequer asking what margin of error is officially regarded as acceptable for the 2001 Census in (a) inner city areas and (b) rural areas. (92496) I published the margins of error, also known as confidence intervals, when the first results from the 2001 Census were released on 30 September 2002 for England and Wales (and inner city and rural areas.) For England and Wales as a whole, the 95 per cent confidence interval (the range within which the true population would fall for 95 per cent of the times the sample was repeated) was plus or minus 0.2 per cent or 104,000 people. For local authority areas the percentage margins of error are larger, ranging from 6.1 per cent in Luton to 0.6 per cent in Dudley, East Dorset and Redcar and Cleveland. All Census population estimates have been subject to rigorous quality assurance which followed an agreed strategy which had been the subject of wide consultation with census users. The population of each local authority by age and sex was considered in a consistent and detailed manner—this involved comparison against diagnostic ranges derived from rolled-forward population estimates and aggregated administrative sources (such as Birth Registration and Pensions data). The quality assurance process means that the census figures are the best estimates we can provide for the population and they are all within acceptable confidence intervals. 425W Of importance also to confidence in census results is that we can explain changes and unusual results. There is insufficient cohesion between the measures of the various contributors to intercensal change in the United Kingdom. This is not acceptable to many users. We are reviewing at present our methods of measuring external migration, and I have initiated a review to identify what we can do to estimate more effectively the population between censuses in areas of large change from sources we cannot measure well.